1. Volume 142, Issue 1, Pages 63-70.e5 (January 2012)
Maintenance of Remission Among Patients With Crohn's Disease on Antimetabolite Therapy After Infliximab Therapy Is Stopped 
Edouard Louis, Jean–Yves Mary, Gwenola Vernier–Massouille, Jean–Charles Grimaud, Yoram Bouhnik, David Laharie, Jean–Louis Dupas, Hélène Pillant, Laurence Picon, Michel Veyrac, Mathurin Flamant, Guillaume Savoye, Raymond Jian, Martine DeVos, Raphaël Porcher, Gilles Paintaud, Eric Piver, Jean–Frédéric Colombel, Marc Lemann 
Gastroenterology 
Volume 142, Issue 1, Pages e5 (January 2012)
DOI: /j.gastro
Copyright © 2012 AGA Institute Terms and Conditions

2. Figure 1
Flowchart of the patients in the study. (A) Disposal of all included patients up to relapse, withdrawal, 30 months, or closing date. (B) Disposal of relapsing patients up to third infliximab re-treatment infusion.
Gastroenterology  , e5DOI: ( /j.gastro )
Copyright © 2012 AGA Institute Terms and Conditions

3. Figure 2
Kaplan–Meier time-to-relapse curve of the 115 included patients. The median ± SE follow-up time was 28 ± 2 months. There were 52 patients with confirmed relapse. The median time to relapse was 16.4 months.
Gastroenterology  , e5DOI: ( /j.gastro )
Copyright © 2012 AGA Institute Terms and Conditions

4. Figure 3
Kaplan–Meier time-to-relapse curves according to multivariable models and scores generated through the Cox model using the multiple imputation method. (A) According to a complete model: with this model (Table 2), the subgroup of patients presenting 3 deleterious prognostic factors or less corresponded to zero to one relapse over 1 year among 22 to 25 patients, depending on imputations. (B) According to a simplified model without infliximab trough levels and endoscopic data: with this model (Table 2), the subgroup presenting 2 deleterious prognostic factors or less corresponded to 4 relapses over 1 year among 32 to 35 patients, depending on imputations.
Gastroenterology  , e5DOI: ( /j.gastro )
Copyright © 2012 AGA Institute Terms and Conditions

5. Figure 4
Profiles of the patients corresponding to the lower risk stratum defined according to the simplified multivariable model. Depending on the gender and the previous surgical history, a variable number of laboratory parameters associated with a lower risk of relapse are requested to be part of the lower risk stratum. For these laboratory parameters, the thresholds for a lower risk of relapse are as follows: leukocyte count ≤6 × 109/L, hemoglobin level > 145 g/L, hsCRP level <5 mg/L, and fecal calprotectin level <300 μg/g.
Gastroenterology  , e5DOI: ( /j.gastro )
Copyright © 2012 AGA Institute Terms and Conditions

6. Supplementary Figure 1
Kaplan–Meier time-to relapse curve of the 115 included patients when using investigators' declaration of relapse, even in the absence of relapse confirmation.
Gastroenterology  , e5DOI: ( /j.gastro )
Copyright © 2012 AGA Institute Terms and Conditions

7. Supplementary Figure 2
Distribution of the concordance probability estimate (CPE) obtained from 2000 bootstrapped samples for the complete and simplified models when expressed as 4 risk group strata. CPE estimate and 95% confidence interval on the original sample are displayed in red.
Gastroenterology  , e5DOI: ( /j.gastro )
Copyright © 2012 AGA Institute Terms and Conditions